ORGANISMS, BRAINS AND THEIR PARTS
UB PHILOSOPHY OF BIOLOGY CONFERENCE
David B. Hershenov
1
I. Introduction
The brain has been described as the organ of thought. In the 18th century, Pierre Cabanis
notoriously claimed that “The brain secretes thought as the liver secretes bile.” For some reason, the
19th century materialist Karl Vogt believed the point needed to be made even more emphatically so
he declared: “The brain secretes thought as the stomach secretes gastric juice, the liver bile, and the
kidneys urine.” Countless neuroscientists make claims like the mind is the brain, or the mind is the
functioning brain, or the mind arises from the functioning brain. Fairly typical is a quote like the
following: “Even though it is common knowledge, it never ceases to amaze me that all the richness
of our mental life – all our feelings, our emotions, our thoughts, our ambitions, our love lives, our
religious sentiments and even what each of us regards as his or her own intimate private self – is
simply the activities of these little specks of jelly in our heads, in our brains.There is nothing else.”
(Ramachandram - A Brief Tour of Human Consciousness
If the brain produces thought, the question then arises: “Does the brain think?” It’s difficult to
see why it should not be able to think. But if human organisms think and their brains think, given that
they are not identical, are there then two thinkers where we had assumed just one? How can any of us
be sure that we are the thinking organism and not the thinking brain? Was this paper co-authored? Or
are these pseudo problems, a result of harmless sophistry from a recently tenured academic
philosopher with perhaps too much free time who might be better served by a greater teaching load
and more committee work? Alas, I do not believe that the problem of a thinking brain is trivial
sophistry nor “language gone on a holiday.” In fact, I am a bit worried that the solution to the
problem will necessitate a radical revision of commonsense ontology and scientific anatomy. After
surveying the relevant alternatives and finding them wanting, I will tentatively and rather reluctantly
endorse an eliminativist stance towards any organ that could be a candidate for being a thinker and
any proper part of the organism that includes that organ.
2
II. Thinking “In Virtue” of the Brain
A common response to the problem of there allegedly being a thinking brain embedded
within the thinking human organism is to claim there aren’t two thinkers but that the human
organism thinks in virtue of its brain. Yet what exactly is this in virtue relation and how does it help
reduce the number of thinkers? It is often illustrated by analogy: it is said that we think in virtue of
our brain just as we see in virtue of our eyes. Unlike our touching the floor in virtue of our feet
literally touching the floor, it is not the case that eyes literally see. You and your eyes don’t make
three seeing things. However, this alleged analogy of seeing in virtue of the eyes won’t rid us of an
unwanted thinker. The reason why the virtues of the “in virtue” relation are less than they might
initially appear is that the brain may be able to think in a self-sufficient way that distinguishes it from
eyes that can’t see by themselves. Or at least philosophers in their thought experiments seem to
accept such a possibility and noted neuroscientists back them up. Hilary Putnam writes of brains in
vats and the neuroscientist Antonio Damasio supports the possibility of such brains thinking, with
some qualifications. Sydney Shoemaker hypothesizes brains being transplanted and thought
continuing throughout the transplant procedures. Eric Olson, Peter van Inwagen and Trenton
Merricks write of organisms being pared down to the size of their brain, and the two most prominent
neuroscientists in the debate over the brain death criterion, James Bernat and Alan Shewmon,
maintain that there could be brain-sized organisms. They hypothesize that thought could continue
even if the rest of one’s organs are removed from one’s brain. Without artificial support, the thought
would be very short-term, but our philosophers and neuroscientists believe the requisite support can
be provided.
Some in the audience might suggest that the thinker or subject of thought in such bizarre
cases is not the brain but the composite of the brain and supporting machinery. Thus the brain is not a
thinker but something else thinks in virtue of contributions by the brain and other entities. Denying
that the composite is the thinking entity is easier for those who maintain (like van Inwagen, Olson,
3
Merricks and Bernat) that the organism could become reduced to the size of a brain and yet remain a
living organism. We generally are not tempted to say that a normal-sized human organism connected
to high tech equipment in the hospital ICU has such equipment as parts or itself is part of a larger
thinking object composed of it and the inorganic equipment. We don’t say this even in extreme cases
of Guillain-Barre syndrome which mimics brain death except for detectable cerebral activity. It is
only the hospitalized organism that thinks, though its life depends upon entities external to it.
Likewise, a brain generically dependent on the vat’s machinery would still count as capable of
thinking. It would be the subject of thought, not some cyborg composed of it and the high-tech
machinery. Perhaps if the human organism can’t be reduced in size to that of the brain and still
considered an organism, or if this can’t be considered a case of two organisms resulting from fission,
as Shewmon and I have suggested, then a stronger case can be made for the brain not being capable
of thought but merely being a vital part of a thinking being whether that thing be a full size human
being or a vat/brain composite. (Incidentally, if this is so, then epistemologists should change their
skeptical question to “how do we know that we are not brain/vat compounds”?)
So what can be said in favor of the brain-size organism? I will let Olson argue the case
though similar discussions can be found in Shewmon and Bernat’s more biologically detailed papers.
Olson writes in the Human Animal (pp. 133-34):
In fact, it seems likely that the head would behave as a living organism if it were
attached to a life-support system. The head would be able to regulate its metabolic
rate and wake-sleep cycle. It would retain its muscle tone (even if no consciousness
were present). Its pupils would open according to the amount of light hitting the
retina and the lens of the eye would focus. And so on…Part of what makes something
a living organism, I suggest, is its capacity to coordinate and regulate its metabolic
and other vital functions. A living organism may be prevented from carrying out
those functions…the instruction that my brainstem sends out to the rest of my organs
may not arrive at their intended destinations, or they may nay not have the effects
they are designed to have (“intended” and “designed” by evolution, not
4
consciously.) Nevertheless, the control and coordination mechanisms are
intact…consider what would happen if we turned off the artificial life support systems
in each case. If we switched off the heart-lung machine that the detached head needs
to survive, the head would behave like a dying organism. It would continue to
regulate its internal metabolic activities in its characteristic way until the “control
centers” shut down owing to lack of oxygen. After a few minutes death, with its
characteristic symptoms – lack of coordinated activity, decay and so forth – would
occur.
III. Other Undetached Thinking Parts
Even if a detached but functioning brain is an impossible philosopher’s fantasy – that has
also carried away some neuroscientists who perhaps should have known better - a related problem of
an embedded thinking entity remains for it is undeniable that a human organism can survive the loss
of many of its parts. Whatever is the smallest thinking remainder is something that earlier would
seem to have been an embedded object and thus would pose the same kind of problems as did the
thinking brain. So the problem will arise with arbitrary undetached parts, even if it doesn’t with the
brain. Moreover, it isn’t that easy to dismiss such parts from our ontology. To borrow Olson’s
example, if there exist such things as hands, then one would expect that there also exist hand
complements. Since these hand complements are composed of every part of the human being but
their hands, they should be able to think. And they shouldn’t lose the capacity to cognize merely by
having something attached to them. Nevertheless, the puzzle is most interesting with the brain or
head since these are not always considered arbitrary undetached parts but well established parts,
respectively, of anatomy and folk ontology. So I will continue to suppose that a pared down brain can
think.
IV. Maximal Thinkers
Some philosophers (such as Michael Burke) dismiss the problem of embedded thinkers by
claiming it is part of the very idea of thinking that no thinker can be a proper part of another thinker.
First of all, understanding “maximal” in this way fails to actually guarantee that the human organism
5
thinks. It only entails that if the organism did think, then it would not have a thinking part. So it
doesn’t rule out that a part thinks – such as the brain - and the whole does not. Of course, one can
tinker with the definition of “maximal thinker” in order to make it more likely that the organism is
the maximal thinker by claiming that the maximal thinker is a thinking being that has neither proper
parts that think nor is a proper part of anything that thinks or fails to think. This approach would be
on firmer ground if we could be confident that organisms are not parts of anything (a society with
collective intentions, eco-system, universe etc.), nor could come to be a part of a thing if we were
attached to something in a way that satisfies a plausible compositional principle. Unfortunately, this
redefining of “maximal” seems to just be stipulating away the problem. I doubt that any explanation
is forthcoming for why something can think but later loses the ability to do so merely by being
attached to something else and becoming a part of that larger object. So if there could be thought
when all of a person’s parts but the brain are removed, why couldn’t the brain think when it earlier
was attached to a person’s other parts, or later when similar parts are restored?
V. Little People
Thus if our brain can think, and our organism can think, there appear to be one too many
thinkers. Listeners may no longer be confident that this problem could be “resolved” by denying that
one’s brain has the capacity for producing thought by itself and thus being a subject of thought.
There are two other options that have been pursued in the literature. Some have argued that one
actually doesn’t possess a brain; others have identified themselves with their brains. The strategy of
claiming to be brainless is likely to be met with derisive quips that the advocate of the view has
uttered a self-verifying proposition. But sarcasm and incredulous stares are not the fate of just the
eliminativist. Similar treatment will be bestowed upon those who identify themselves with their brain
and claim that persons are but a small part of the human organism. At least, the eliminativist about
brains doesn’t claim we are only a few inches and pounds and have never been directly observed nor
touched. While it is admittedly good news that the identification approach guarantees that none of us
6
are really, strictly obese or ugly, it is a shame that this has to be achieved by our neither having
bodies nor faces.
Strange as the identification strategy may initially seem, there are prominent philosophers
who defend it (Hud Hudson, Michael Tye, Jeff McMahan, Ingmar Persson, Roland Puccetti, and
Roderick Chisholm). Some claim we are 3D brains or 3D parts of such brains, others claim we are
4D brains. Some have done so to avoid the extra thinker, others for different reasons. 1) McMahan
and Persson describe two-headed conjoined twins, real and fictional. The thought realized by each of
the two brains would be inaccessible to the other so there is a Neo-Lockean pull to say that there are
two distinct persons rather than one person cut off from herself. And since there is only one organism
in such scenarios, organisms can’t be persons on pain of violating the transitivity of identity. Split
brains and dissociative disorders also motivate some to claim we are just part of an organism on the
basis of the same Neo-Lockean reasons about psychological unity. 2) A second reason philosophers
might be attracted to this strange downsizing view is that they don’t see how the brain or certain parts
of it could be denied the capacity to think. The brain is seen in some sense as the minimal thinker,
nothing else is needed or is directly involved in the production of thought, and this makes it the best
candidate for being the thinker of our thought. And as Chisholm rightly insists, if anything strictly
thinks one’s thoughts, one should be identical to that entity rather than one that derivatively thinks in
virtue of something else strictly thinking. 3) The claim that we are our brains receives further support
from the brain transplants and body switching thought experiments that are so popular in the
metaphysics of personal identity literature. 4) A fourth reason to believe we are our brains is that one
believes mental states are brain states. That makes it sound like the brain would be the possessor of
those states and thus the subject of any thoughts.
Although I am attracted to the prospect of neither being strictly bald nor having a beer belly, I
am otherwise not a fan of the identification of the brain and person. There are a number of problems
such as doing justice to our intuitive account of the persistence conditions of both brains and persons
7
if we identify them, and also of trying to cash out the idea of being directly involved in thought. But I
just want to point out that there still looms the problem of the thinking organism. It is hard to believe
that there are organisms possessing our brains that don’t think. McMahan seeks to downplay the
organism thinking by comparing it to a car that is only derivatively noisy because its horn is noisy.
Just as there aren’t two noisemakers, so there really wouldn’t be two thoughtmakers. But matters are
not as innocuous when thought is involved. Imagine that the brain-sized person insisted “I am
essentially a person.” The organism thinking a similar thought in virtue of possessing the same brain
would have derivatively said something false.
Eric Olson points out that there would also be an epistemic problem. The thinking organism
doesn’t know whether it is the person or the organism if it thinks qualitatively the same thoughts with
the same organ as the much smaller person. An additional puzzle concerns why the organism isn’t a
person since it is self-conscious and apparently can think about the past and the future in the way that
the Lockean tradition maintains is characteristic of personhood. Locke famously wrote “We must
consider what person stands for: which, I think, is a thinking intelligent being, that has reason and
reflection, and consider it self as it self, the same thinking thing in different times and places.” The
organism has the capacities of the person but just lacks the persistence conditions of persons. Olson
labels this the personhood problem.
There is an additional difficulty with the claim that the brain is strictly the thinking being and
the organism only derivatively thinks. If Olson, van Inwagen and Merricks are right that the human
organism could live pared down to the size of the brain, then it would be composed of every part of
the brain and nothing else. It is quite odd that it would then only be derivatively a thinker. This
queerness could be reinforced if it were metaphysically possible for a functioning brain to be made
first and then head, neck, trunk and other appendages added later. It is very hard to explain why at
the early stage only the brain would be strictly the thinker. It seems that there would be two bodies
strictly thinking the qualitatively same thoughts. If one believes that the minimal thinking being is
8
just part of the brain and thus smaller than the maimed brain-size organism, one can just reproduce
the problem with the brain and its minimal thinking part rather than the organism and the brain. The
brain can survive the loss of some of its parts and become reduced to the size of the minimal thinker.
These two cases presents us with the familiar puzzle of spatially coincident objects: two things
physically the same and with the same relations to the environment, somehow end up with different
modal qualities, and if an unwanted extra thinker is to be avoided, an account must be given of why
one of the two physically indistinguishable entities can think.
VI. Noonan’s Pronoun Revisionism
Harold Noonan doesn’t think the abundance of thinkers is an insurmountable problem. He
offers the three dimensionalist confronting spatially coincident and embedded thinkers an account of
linguistic revision designed to serve his favored four dimensionalist approach which allegedly avoids
the above problems. He separates the thinkers of our thoughts from the referent of “I.” The idea is
that the first person pronoun doesn’t refer to all those who think or utter it but only to the person who
is distinguished by his persistence conditions. So neither a brain that is part of the organism, nor an
organism spatially coincident with the person, nor a perduring organism overlapping a perduring
person, nor a stage that is part of a four-dimensional worm refers to itself. The first person pronoun
“I” refers not to the thinker of the thought as we English speakers may claim to have been taught, but
to the individual that is a person. Noonan insists that it is an a priori truth that the reference of the
first person pronoun is a person and that metaphysical arguments can be made to show that the
person is the entity that has the identity or persistence conditions of a thinking being rather than those
of an essentially living being. As a result he insists that the organism, brain and the person can think
truthfully and knowingly when they think “I am the person” because the referent of all the thinkers is
the person.
I don’t think Noonan should have offered a solution based on just changing our
understanding of how we speak. Rather he should make a claim about how we think, or if you like,
9
about the language of thought. His claim shouldn’t be just that the word “I” refers only to the person,
but that the organism doesn’t have the capacity to refer to itself in a way that is characteristic of a
self-conscious person. It is the nature of the person to be capable of thinking about thoughts that it
knows to be its own. The organism cannot think about its thoughts in this way. And the same can be
said of the brain. So it isn’t that a certain meaning of “I” in Kaplan’s sense of character has been
learned rather than a different one, but that the organism and the brain are psychologically incapable
of reflecting on their own thoughts as their own. If the difference between the organism and the
person’s referential abilities were just a matter of conventional linguistic usage then we could just as
well stipulate that “I” refers to the organism, and the organism could then refer to itself in a way the
person couldn’t. However, such a move should merely take away the person’s ability to refer to itself
by that word rather than to self-consciously think about itself. And while such a stipulation would
allow the organism to refer to itself by “I,” it should not bestow upon the organism the capacity to
think that it is identical to the referent of “I.” Otherwise, the epistemic problems return.
So Noonan’s idea should be one about the nature of thought and not just about the referents
of our natural language. While the organism and brain have second-order thoughts, thoughts about
thoughts, they are not thoughts about themselves qua organism or qua brain. They have a very odd
sort of direct, immediate, non-inferential idea of what someone other than themselves is thinking, but
they don’t have self-knowledge. Organisms, as well as brains, can think about themselves in the way
John Perry’s sugar spilling shopper can by some description. There may even be a sense in which
they think about their own thoughts since they are co-generating the person’s thoughts and aware of
those thoughts. But they aren’t aware of those thoughts as thoughts of their own. They can’t conceive
of themselves as thinking beings, as entities with thoughts about themselves. Thus there is no
personhood problem as envisioned by Olson: two or more entities with the very same mental
capacities but only one is a person due to its distinct persistence conditions. And since the first
10
person pronoun refers to the person, there is no epistemic problem as such questions can’t be raised
by the non-persons.
What we should demand of such an account is an explanation of why the organism or the
brain can’t refer to themselves. I doubt that a compelling answer is forthcoming. Why should the
organism be unable to do what the person can if they are physically the same and causally related to
the same environment? If persons learn that the first person pronoun “I” refers to themselves, why
don’t organisms learn the same lessons and thus refer to themselves? Likewise for the thinking brain:
If the person or organism can be pared down to the size of the brain, why can’t the brain selfconsciously think like the person or the organism at that time? And even when the organism or
person hasn’t been reduced to the size of the brain, there is still quite a mystery about what the brain
is missing that would prevent it from having the same mental capacities. It would seem to share the
relevant physical makeup of the organism or person and would be causally related to the same
environment, though a few steps removed. So while pronoun revisionism, at least when supported by
a language of thought construal, can solve the personhood problem and prevent the epistemic
problem from arising, it does leave a mystery.
Some in the audience might insist that it is just a brute fact that the person has the ability to
think and the brain and organism do not, assuming that the organism is distinct from the person. But
this is an unattractive move. “Going brute” would be more appealing if we were wondering why
material simples like electrons can’t think but immaterial simples like souls allegedly can. With
simples, there is nothing to explain. Appeals are just made to the nature of things. There are not parts
with properties and standing in relations to other parts that can explain the macro phenomena in
question. But since organisms and brains are complex composite objects, one would think the
difference in their cognitive capabilities should have an explanation in virtue of the properties of their
parts.
11
VII. Coming to Terms with Being Brainless
Running out of options, the audience might want to join me in seriously considering
eliminativism now that a look of puzzlement has perhaps replaced the incredulous stare. This is van
Inwagen, Merricks and Olson’s solution to the problem of too many thinkers - and the one that I
accept though reluctantly. They eliminate brains from their ontology, identify persons with
organisms, and reject the metaphysics of temporal parts. In fact, they eliminate all undetached parts
that are large to enough to subsume the brain and apparently be capable of thought such as hand
complements. Independent support for this approach is provided by van Inwagen’s compositional
principle that Xs compose a Y iff they are caught up in a life. This is not as arbitrary (and disjunctive)
a compositional principle would have to be to include cerebrums, heads and hand complements.
Olson, van Inwagen and Merricks suggest that we speak of atoms arranged brain-wise that
don’t compose anything where others refer to brains composed of atoms. (These are philosophical
atoms, simples, not the composite atoms of the physicist.) Ordinary object language gets paraphrased
in a way in which is in some sense close enough to true and can serve our purposes. Falsely claiming
that there are brains when there are only atoms arranged brain-wise, is a lot different from falsely
claiming that there are any unicorns for which no satisfactory paraphrase is available.
VIII. Tu Quoque?
There are various costs of paraphrase but I don’t have the time to go into them here. Perhaps
more worrisome for the Olson and van Inwagen-style theorist advocating a sparse ontology of just
organisms and simples is the possibility of conjoined twins that share a cerebrum but no other organs
(or rather atoms arranged organ-wise.) There was an actual case of conjoined fetal twins which
shared a cerebrum but not a brain stem nor any other vital organs involved in the life processes
thought to individuate organisms. Since they each had their own brainstems controlling different life
processes there would have been distinct organisms on Olson and van Inwagen’s accounts of
organism individuation. But each organism would be the subject of thought, thinking thoughts that
12
appear to be qualitatively alike. Although the just described twins died when late term fetuses,
perhaps then capable of only minimal sentience, if they had lived, the shared cerebrum might have
brought with it the problem of too many thinkers and the epistemic bewilderment that Olson argues
his opponents suffer. While such cases of conjoined twins are rare, we can conceive of a possible
world in which they are not uncommon. And claims about the nature of human persons and human
organisms and their relationship should apply in every metaphysically possible world. We don’t want
an account of the relationship of organisms and their brains to work just for our actual world.
So audience members might be thinking that if even the Olson/van Inwagen-inspired
ontology must accept an extra thinker then they and their opponents are in the same leaking boat and
thus there is no reason to advocate a sparse ontology. (Doubts about de re vagueness might also give
this result.) We should just accept thinking persons, thinking organisms and thinking parts and rely
on pronoun revisionism to take the sting off this. However, I think rival accounts (except for
immaterialism – of some varieties of Cartesian and emergentist dualism) fare even worse in this
cerebrum sharing case. My materialist rivals must posit more thinkers in the conjoined twin case as
well as in normal scenarios. Furthermore, the van Inwagen-Olson ontology doesn’t get stuck with
embedded thinkers nor is there any violation of supervenience and a modal mystery as there are in
cases of spatially coincident thinkers, so it doesn’t share that weakness with most of its theoretical
rivals. Thus I will bite the bullet and just accept the fact that I am unable to account for any selfreferential thought in such cerebrum sharing twins. So in the case of the two organisms sharing a
cerebrum, there would either be a failure of reference or, if reference succeeded, it would not be
accompanied by the ability to ever know, when connected, which twin one was.
My hunch, though not necessarily my hope, is that the atoms arranged brain-wise in the
audience will not be able to make any contribution to a better solution here or in any of the other
scenarios of extra spatially coincident, embedded or overlapping thinkers.
13